---

Study of resources changes in previous years could be useful in the planning and optimal using of resources to control inappropriate changes. Because land use changes occur on large-scale, remote sensing technique is a useful and valuable tool for monitoring the changes. The aim of this research is land  cover changes detection in a period of 32 years in Pishva town with using remote sensing technique .First TM, ETM and OLI images for the years 1986, 2002 and 2018 were collected respectively and after geometric and radiometric corrections, images were classified by using maximum likelihood classification methods. Kappa and overall indexes were used to calculate classification accuracy. Results showed in past 32 years, bare land and irrigated land have decreased while residential and greenhouse areas have increased. Classification accuracy showed that OLI, ETM and TM sensors have high accuracy respectively with kappa 0.96, 0.80 and 0.76 and also overall indexes of 97.56, 86.54 and 86 percent. Based on results, in the first period (1986-2002) 27.6%, in the second period (2002-2018) 29.60% and in the third period (1986-2018) 31.8% of area land cover have been changed. Results showed land cover changes in the area is related to climate changes like low precipitation, drought and social condition like population and food need increasing and economic condition like high production and efficiency.
 

---

Human interventions in natural areas as a change in land use have led to a domino effect of anomalies and then environmental hazards. These extensive and cumulative changes in land cover and land use have manifested themselves in the form of anomalies such as the formation of severe runoff, soil erosion, the spread of desertification, and salinization of the soil. The main purpose of this study is to reveal the temperature inductions of the land cover structure of Lorestan province and to analyze the effect of land use changes on the temperature structure of the province. In this regard, the data of land cover classes of MCD12Q2 composite product and ground temperature of MOD11A2 product of MODIS sensor were used. Also, in order to detect the temperature inductions of each land cover during the hot and cold seasons, cross-analysis matrix (CTM) technique was used. The results showed that in general in Lorestan province 5 cover classes including: forest lands, pastures, agricultural lands, constructed lands and barren lands could be detected. The results of cross-matrix analysis showed that in hot and cold seasons, forest cover (IGBP code 5) with a temperature of 48 ° C and urban and residential land cover (IGBP code 13) with a temperature of 16 ° C as the hottest land use, respectively. They count. In addition, it was observed that the thermal inductions of land cover in the warm season are minimized and there is no significant difference between the temperature structure of land cover classes; But in the cold season, the thermal impulses of land cover are more pronounced. The results of analysis of variance test showed that in the cold period of the year, unlike the warm period of the year, different land cover classes; Significantly (Sig = 0.026) has created different thermal impressions in the province. Scheffe's post hoc analysis indicated that this was the difference between rangeland cover classes and billet up cover.